Ensemble Simulation of Anthropogenic and Biomass Burning CO2 and CO in CAM-Chem

Anthropogenic combustion and biomass burning has significant impacts on air quality, climate, ecosystem, agriculture, and public health at both regional and global scales. CO₂ and CO are the key combustion signatures that can be used to characterize combustion processes, and their relationships also provide a unique perspective on emission sources. To reach a better understanding of anthropogenic combustion and biomass burning, modeling atmospheric CO₂ and CO, as well as their relationships is critical. In this study, we conduct ensemble simulations of atmospheric CO₂ and CO (and associated tagged tracers) in CAM-Chem. As CO₂ fluxes from land and ocean is not the main focus of this study, we use prescribed surface fluxes from previous CO₂ forward/inverse modeling studies constrained by observations. The focus of this study is CO₂ and CO emitted from anthropogenic combustion processes in megacities, and biomass burning processes in major biomass burning regions (such as Amazon, equatorial southern Africa, Indonesia, Canadian boreal forests, and Siberian boreal forests). CO₂ and CO emitted from megacities provide information on anthropogenic combustion activities, while CO₂ and CO emitted from biomass burning regions provide information on fire characteristics such as modified combustion efficiency (MCE=CO₂/(CO₂+CO)). We compare ensemble model simulations across the aforementioned regions to quantify differences in characteristics of anthropogenic combustion activities among different megacities, and differences in fire characteristics across biomass burning regions (such as flaming or smoldering). This study provides an opportunity to confront these ensembles with multiple constraints including enhancement ratios (which are useful to assess relationships between urbanization and anthropogenic emissions), and smoke index (defined with satellite observations of NO₂, CO, AOD from OMI, MOPITT, and MODIS to infer combustion efficiency). In addition, we investigate error propagation through ensemble perturbations and transport errors in CAM-Chem as well as the consistency in CO₂ and CO abundance. We also evaluate CAM-Chem ensemble results with satellite observations using CO from MOPITT, and CO₂ from OCO-2 and GOSAT. In addition to satellite retrievals, we also evaluate our results with ground site observations (e.g., TCCON), airborne measurements (e.g., KORUS-AQ, and DISCOVER-AQ), measurements of isotopes, and other model products (such as CAMS from ECMWF). Through comparisons with observational data, we gain insights on the representation of combustion processes in CAM-Chem, and emission inventories for both anthropogenic combustion and biomass burning.